Evaluating the Minimal Clinically Important Difference of the King's Sarcoidosis Questionnaire in a Multicenter Prospective Study

Rationale: Improvement of quality of life (QoL) in patients with sarcoidosis is an important goal of management. The King's Sarcoidosis Questionnaire (KSQ) and Patient Global Assessment (PGA) are instruments that have been used in sarcoidosis.Objectives: We defined the minimal clinically important difference (MCID) as the within-patient clinically meaningful change threshold and determined the MCID of KSQ general health (KSQ GH), KSQ lung, and PGA using both anchor and distribution methods. The discriminatory properties of these MCIDs relative to other QoL instruments were then determined.Methods: Patients with sarcoidosis recruited from six centers in the United States were prospectively studied. Initially and at 6 months, patients completed a series of QoL questionnaires, including the St. George's Respiratory Questionnaire (SGRQ), Short Form 36 (SF-36), Fatigue Assessment Scale (FAS), Sarcoidosis Assessment Tool (SAT), KSQ, and PGA, and spirometry. For the anchor method, receiver operator characteristic curves were used to determine the MCID for improvement or worsening. The distribution method using half of the standard deviation was calculated for KSQ GH, KSQ lung, and PGA.Results: Of the 325 patients enrolled in the study, 271 completed the 6-month evaluation. At 6 months, approximately half of patients were worse and 30% were improved based on previously established MCID values for the SGRQ, SF-36, and FAS. There were no discordant cases. There were significant correlations between the KSQ GH, KSQ lung, and PGA and most parameters assessed. The best correlations were with the SGRQ, SF-36, and FAS, which have established MCID values. Using anchor analysis, we found that most of the domains of SGRQ and SF-36 were able to determine the significant MCIDs for all three variables. These MCIDs were similar to those determined by the half least square method. We propose an MCID of 8 for the KSQ GH, an MCID of 4 for the KSQ lung, and an MCID of 2 for the PGA because these values captured >90% of parameters studied. These MCID values discriminated between changes in other QoL instruments.Conclusions: The determination of MCID values for KSQ lung, KSQ GH, and PGA may prove useful for clinical practice as well as clinical trials.